Hacker News new | ask | show | jobs
by skyfaller 649 days ago
I would have been more interested in Cohost if they weren't actively hostile to federation. Their eventual shutdown always seemed inevitable to me, and without federation + account portability, there would be no easy offramp when they failed.

You could say the same of almost all corporate social media, and I am phasing them all out too.

2 comments

Federation wouldn't have meaningfully solved their revenue problems, though. It would have just been a massive engineering timesink.
It could have helped their revenue problems indirectly if it meant being able to access more users and content, not only the people who use and enjoy Cohost. People love to talk about network effects, and something like the fediverse means you can start with however many users there are on the fediverse on day one of turning on federation.

For me, lack of federation meant they were dead on arrival, just like every Google product can't be trusted to keep existing (and I would never start using a new Google product). Any individual fediverse project could shut down, but the fediverse continues and is probably unkillable like IRC or RSS at this point, even if it's not as widely used as one might like.

> It could have helped

Narrator: It could not have helped, it would have mostly just made them even more burnt out, because they (as they said in the post) were already on call effectively 7 days a week and have been since the site started.

> the fediverse continues and is probably unkillable like IRC or RSS at this point

Who cares? When an IRC channel, or phpbb forum I use closes down, or an RSS feed I used to read goes dark -- I do not think about the wonderful IRC protocol and how ossified it is or about the sanctity of DNS. And these things happen regardless of what the underlying technology stack are. I think "It really sucks that my community and the people on it are gone now." A domain that doesn't resolve and an empty mastodon instance are effectively the same.

The users who use the platform are why the platform is valuable, not because it uses some dumb particular tech stack.

This comes down to a question of what kind of service you want.

If you want a way to "access more users and content", you can already make an account on a Fediverse instance and access those users and content there. What does Cohost offer by being yet another instance with custom software?

If the vast majority of the content I'm accessing is from other instances/services, why would I pay Cohost money for the privilege of having them act as an intermediary?

One specific problem with federation/decentralization is that you've now decentralized moderation and given up control over your service's culture entirely. This has some upsides and some downsides, but it again puts you into a weird position: If your moderation and culture position are identical to Mastodon's, why do you need to exist? If you're just another Mastodon instance, why would anyone give you money?

When I post on HN or Twitter or Tumblr or Cohost, I at least have a good sense of how the service is moderated, what the rules are, etc. When I post on decentralized services, at any point I can discover that oh, this random instance defederated the one I was unlucky enough to sign up on, so half of my mutuals can't see my posts anymore. Don't worry, you can just migrate your account to a different instance with different rules, and hope IT doesn't get defederated! And because each instance has different rules and culture, you get to look forward to people from other instances complaining that you aren't complying with their rules. It's messy! It's not fun!

Of course the rules on those services I mentioned aren't necessarily going to suit everyone's tastes. But I think that's good - a social media service doesn't need to be For Everyone to be successful, and being For Nobody is a horrible outcome.

At the end of the day I see the appeal of defederation but it simply doesn't make sense as a way to spend your engineering dollars if your goal is to be profitable. It Doesn't Make You Money.

So you are saying that you would pay for an account for Mastodon or Lemmy?
I am an admin / mod for a Mastodon server that is supported by donations from the community. I have not put my own money in the pot for a while, because I'm donating a lot of time to run it.

But I would pay if I had to, and I am considering paying for a GoToSocial server to experiment with an allowlist network:

https://gotosocial.org/

https://codeberg.org/oliphant/islands/src/branch/main/ion

My server: https://jawns.club

Our finances: https://opencollective.com/jawnsclub

We're currently paying for managed Mastodon hosting on https://masto.host/

How do you think your community would react if you switched to a "everyone pays a little bit every year" model?
I'm sure we would lose some people if we switched to mandatory payments. Since this is already a pretty small community (dashboard currently says 171 active users), I wouldn't care to experiment, since we'd risk losing the critical mass necessary to have an active local timeline, which for me is a major reason to run your own server. I'd also hazard a guess that many people go inactive for a while, and then check in again randomly when they need more social media in their life for whatever reason, and mandatory payments might interfere with that movement in and out of inactivity.

Finally, I think pay what you want is better so long as it works, since it doesn't exclude people who don't have money, but do contribute to the community in other ways. The only real reason to move away from pay what you want is if it doesn't pay the bills, and our finances are fine for now.

I can see a place for mandatory payments if you're providing extra services at a steeper price, such as paid moderation, but I think the number of people willing to pay what moderation at a living wage actually costs... is rather small. Perhaps if we made moderation more efficient, e.g. sharing moderation decisions between servers, paid moderation could become more affordable by splitting the cost between more users, but there are several problems with that approach... one being that moderation by members of your community is always going to be more clueful than moderation from outside your community.

Do you think it's a fair assessment to say then you are making the same argument I mentioned in the other thread: very few people think that the service of a social media account is actually worth anything, and that this should only be treated as a hobby?

Follow up question: if everyone treats social media alternatives as a hobby, do you think that it has a chance of being a viable alternative to the Big Tech platforms?

Big tech platforms aren't a viable alternative to big tech platforms, just look at the demise of Andreessen-funded Post News, if you're tired of me mentioning Google's failures: https://www.theverge.com/2024/4/19/24135011/twitter-alternat...

I think the profit motive is literally destroying the world given the climate crisis, as well as destroying everything good or useful about the Internet. So I think figuring out how to do things without the profit motive is the only way forward, as impossible as that might seem. Surviving capitalism with that attitude may be challenging, but we don't have a future under capitalism anyway.

I think that if the value of social media comes from the network effects, then it's not a question of whether you can find some core of users who consider it more than a hobby. (Ugh, you want a network for professionals for whom it's their job?!) For social media to reach its maximum value, it has to reach literally everyone, and at that point you're talking about something that ought to be a government service, like the postal service. For smaller networks that are less focused on reach, maybe hobbyists are the ideal providers.

You might be surprised to learn that many people do donate money to their instance of choice. Enough to cover server and admin costs? Probably not. But it shows some willingness to pay.
This is not what I am asking. I am asking if OP would join a commercial provider of Mastodon, where access is only given to paying subscribers.

I am asking because I happen to run one of those (https://communick.com) since 2019.

I was actually searching for a commercial Mastodon host around a year ago. I would have been interested in your service but it did not come up in any search results. I remember checking out Librem One, but the signup process for that was very buggy. Other services were targeted towards people who wanted to host instances themselves, not just have an account.

Even now I can’t find your service in the first page of my search results. I ended up just setting up a recurring donation to Mastodon.social via Patreon.

Your service looks very cool. Just pointing out that it is very hard to discover even for people who are willing to pay, as the GP comment notes and hopefully providing a useful experience report.

Yeah, in my case it gets specially hard because Mastodon's project page does not point to commercial providers, only hosting services.

To be honest though, I think that Mastodon is an evolutionary dead-end. It had a huge head start in the space, but I am reasonably sure that it is not the future of the open social web. I am more inclined to pivot into a multi-protocol client (like Pidgin) and offering ancillary services than trying to pick a champion and invest into promoting it.

> To be honest though, I think that Mastodon is an evolutionary dead-end. It had a huge head start in the space, but I am reasonably sure that it is not the future of the open social web.

Interesting, this was my layman's read too after a while of trying to get into it. To me it felt like it was opinionated but in ways that I disagreed with.

For example, instead of a separate UI for private messages, they just add another option to the visibility selector. From a technical perspective I suppose this is somewhat elegant, but from a UX perspective it felt wrong. I saw a thread where people disagreed with it and the maintainers told them to pound sand. Similarly with quote tweets.

I know you mentioned not trying to pick champions, but are there any alternatives that you thought were particularly interesting? I'm willing to put in a little time to try things out.

The other thing I want to note is I just tried to use the "Set up Auto-Pay" button on Communick and got the error "Could not set up auto pay. Please contact support". I'd love to sign up though.

Probably cheaper to run a mastodon instance, since there are no development costs.